Built for kids,
by kids.

It began with a simple question at the kitchen table: could we make a real game? Not a worksheet, not a tutorial with the answers already filled in — a real one, on a real website, that a friend could open on their phone.

So we started building. The kids bring the ideas, describe how a game should feel, and decide when something is fun or boring. AI helps turn those ideas into working code, and every round teaches a little more about how software actually gets made.

Every game starts as a kid's idea — a sketch on paper, a description shouted from the back seat, or "what if the snake could go through walls?". Some are brand-new inventions. Some are our own takes on the classics, because classics are how you learn what makes a game work.

The kids make the real calls: how fast it should go, what colour the ship is, whether losing feels unfair. Those decisions are the whole point. Writing the code is the easy part now; knowing what to build is the skill worth having.

AI can write code in seconds, which makes it tempting to skip the thinking. We try to do the opposite. Describing an idea precisely enough to build it, spotting when the result is wrong, testing it on a real phone, and fixing it — that is the actual work, and none of it gets automated away.

Along the way there is a lot of ordinary, useful stuff: how a game loop runs, why a score needs to be saved somewhere, what happens when two people play at once, and why "it works on my computer" is never the end of the story.

The arcade grows as the kids learn. Each new game is a bit more ambitious than the last, and older ones get tweaked when someone decides a level is too hard or a colour is wrong.

If your family has an idea for a game, we would genuinely love to hear it — some of the best ones have come from outside this house. Send it over and it might end up in the arcade.